Professional Documents
Culture Documents
Mass Air Flow Sensor: Content
Mass Air Flow Sensor: Content
Page
content:
Pierburg MAF
Bosch MAF
Converting mg/stroke to kg/hour
The Pierburg MAF, VW number 074 906 461, has 6 pins. The wire colors and the connections mentioned to the ECU are from a 66kW AGR and a
81kW AHF engine.
Pin
Description
Wire color
Connected to
Red/green
Signal ground
Brown/blue
Yellow/black
relay 109
Not connected
Power ground
Brown
ground
Output voltage
Purple/brown
The output voltage is proportional to the air flow through the device as can be seen in the curve below.
VW uses several Bosch MAFs for its TDIs. Bosch number 0 281 002 216, VW number 028 906 461 and also Bosch number 0 280 217 121, VW
number 06A 906 461. As far as I know these MAFs are equal. The drawn transfer curve drawn is from a Bosch 0 280 218 018 MAF. I'm pretty sure
this curve is equal to the curve of the previous mentioned MAFs. The wire colors and the connections mentioned to the ECU are from a 66kW ALH
engine.
A Bosch MAF has 5 pins:
Pin
Description
Wire color
Connected to
1
2
Yellow/black
relay 109
Pin
Description
Wire color
Connected to
Ground
Brown
Red/green
Output voltage
Brown/blue
At 900 RPM a 4 cilinder 4 stroke engine makes 900 RPM * 2 strokes/revolution * 60 minutes in one hour = 108,000 strokes. So if the ECU reads
480 mg/stroke at idle, it equals to 108,000 * 480e-6 = 52 kg/hour air flow.
480 mg/stroke is the value you should read with a 1.9L engine during idle when the EGR is closed. One full cilinder filling is 0.48 liter. Air is about
1290 mg per liter. So one could calculate 0.48 * 1290 = 619 mg/stroke. However, the cilinder is not fully filled as it has to suck in the air in a
limited amount of time. Therefore the value is a little lower. (If the turbo would be really working at idle, the air flow could be higher than 619
mg/stroke.)
The output voltage of a Pierburg MAF during idle and closed EGR is about 1.6V. A Bosch MAF reads about 2.2V in this situation. With a closed
EGR you should measure about 480mg/stroke. With a 78% opened EGR you should read about 290 mg/stroke. If the values you find are much
lower, something is blocking the air flow or the MAF has gone bad. The air flow can be blocked by a clogged air filter, a clogged intercooler, a
(partly) closed anti shudder valve, a clogged intake manifold or bad valves. The intake manifold may clog due to soot/oil as exhaust fumes are fed
back into the engine by the EGR and oil is fed into the engine by the CCV. It's also possible that the EGR is stuck open.
Below is a curve of the mass air flow of a 1.9L TDI industrial engine (AFD).
Page content:
VW uses several Bosch MAFs for its TDIs. Bosch number 0 281 002 216, VW number 028 906 461 and also Bosch number 0 280 217 121, VW
number 06A 906 461. These MAFs are equal, except for the length of the housing.
The graph below was measured on a 81kW TDI (engine AFN, after 180,000 km) under WOT (wide open throttle). As can be seen clearly, the
specified air flow is much higher than the actual (measured) air flow. Turbo pressure reached 2000mbar, so turbo seems to be fine. Also the intake
manifold was not fouled by soot and oil.
The circle in the graph shows the 3000RPM WOT point. Here you should read 850mg/stroke MAF (specified and actual). Turbo pressure should be
2000mbar. (not in this graph)
Reading the fault memory of the ECU did not show any errors. Still the owner felt his car was lacking response to the go pedal.
A new MAF on the same engine immediately cured the problem. Power returned, specified MAF and actual MAF are equal again.
The 3000 RPM WOT point shows 850mg/stroke. As can be seen, actual and specified MAF are not always exactly the same. This is normal.
In the graph below another reading with an AHF engine (same MAF as in AFN) is shown. Full throttle acceleration from first till fourth gear
(manual shift). Shift points at 47, 53 and 60 seconds. Specified MAF is 850mg/stroke at 3000 RPM. The actual MAF is also 850mg/stroke: The
MAF is okay.
According to my information the MAF signal wire is brown/blue. (your colors may be different) It is pin number 5 of the MAF connector. So the
output signal from the MAF is connected to the variable resistor. The signal to the ECU can be adjusted. One more resistor is needed on the other
side of the variable resistor (10k ohm). It is connected to ground. (pin 3 of MAF connector brown)
For best results you need a VAGCOM or similar. Measure the MAF with a TDI MAF and adjust the variable resistor such that the same value is
read again with the 2.0L MAF. Also a simple voltage meter will do just fine: measure the output voltage of a TDI MAF and adjust the resistor such
that the same voltage is read on the wire going to the ECU when the 2.0L MAF is hooked up. If everything is okay, the smoke should be gone, and
the engine is just as powerful as it was with the TDI MAF. Only above 4000 RPM or so, a very small power loss may be experienced, but I don't
think anyone will ever notice.
If you don't want to adjust anything, you can use fixed resistors. The picture below shows the same circuit with fixed resistors. The values should be
about right for people experiencing smoke with the 2.0L MAF, but not with the TDI MAF. If the engine is still smoking with this circuit, increase
the 820 ohm resistor to 1k ohm or even larger. If the engine starts to feel sluggish, the resistor is too large.
Fault tolerant
The circuits are designed to handle all kind of fault conditions. Shorting the inputs or outputs to any voltage between 0 and 12V will not damage the
circuits. The supply voltage is 5V, but if you would apply 12V, nothing will go bad in these circuits. Although the circuits here are fault tolerant, I
do not know if the MAFs or ECUs are. The ECU can handle voltages between 0 and 5V for sure, but I'm not sure it will not be damaged by voltages
above that.
Accurate
The conversion is quite accurate as you can see in the measurements below. It's best to use 2% resistors. You can use 5% resistors, but you may
want to check the actual resistance value. Normally 5% resistors are within a few percent, so good enough.
Low cost components
The components used are all very low cost. 0.25W resistors and standard small signal transistors. You can find these at any electronics shop at a
cost of something like one dollar in total for the parts. I used BC547/BC557 transistors, but many small signal transistor will do. (2N2222/2N2907
is okay too) Don't waste your time making a beautiful printed circuit board. Just use an experimental PCB. That really is good enough. The circuits
are not critical on PCB layout at all.
Casing
For the casing it is important that it is water tight. If not, the circuit will corrode and will go bad in no time. The circuits are so small however that it
will fit in a 35mm film container. Just punch small holes in it for the wires and seal it with silicon. The container can be placed next to the MAF.
Intellectual property
You are free to use the information presented here. I do not ask any money for it, but also do not accept any liability. You are not allowed to sell the
schematics or printed circuit boards with these converters with a profit. If you find a circuit valuable and are going to use it, I would be happy to
hear your experience with it.
Page content:
Using a Pierburg Mass Air Flow sensor with an ECU programmed for a Bosch MAF
The schematic is shown here. It basically is an attenuator that also does some level shifting and curve shaping. In the interesting air flow range from
idle to full power (50kg/hour to 400kg/hour) the conversion is very accurate. This circuit has been tested for quite some time in a 81kW TDI. It
work beautifully. No error codes in the ECU (of course not, the ECU doesn't even see the difference between a Bosch and a Pierburg MAF. So why
should it complain when it does get the right signals.)
Using a Bosch Mass Air Flow sensor with an ECU programmed for a Pierburg MAF
The schematic is shown here. It basically is a low gain amplifier that also does some level shifting. In the interesting air flow range from idle to full
power (50kg/hour to 400kg/hour) the conversion is very accurate. This circuit was tested, but only on the lab table. As the conversion is very good
too, there's no reason to suspect that this circuit will not work.
With that conclusion I made a circuit that simply amplified the MAF output voltage with 15% starting from 0 airmass or rather 1 volts:
The circuit is made with a differential pair Q1 and Q2 driven by a current source made of Q3 and Q4. The output stage consists of Q4 with Miller
capacitor C1 for loop stability. The gain of the circuit can be adjusted by R10. Tune R10 all the way to zero ohms for a good MAF and when the
MAF gets older (bad) increase R10. With the bad MAF I had, I had to set R10 to about 600 ohms (about half way) to get the reported airflow back
on target.
Again all bipolar transistors. This to make sure that the circuit will work under any condition. Commercial available integrated circuits are normally
not specified over the automotive temperature range. So they may deviate or even fail when temperatures are outside the specification.
Road test
With the circuit I did a road test. The performance was good. The air mass reported was back to 850mg/stroke at 3000RPM WOT and also a Butt
Dyno (see www.tdiclub.com) showed that the performance was back.
Logging results
First I did a log with the bad MAF:
Pretty amazing, even for me. All loggings were done with the same '98 81kW TDI on the same day.
Conclusion? It seems that even a very old and very bad MAF is still usable and will perform like a new one with a little help of a tiny boost circuit.
How is it connected
The output wire of the MAF (pin 5 of Bosch MAF) is cut and the circuit below is inserted. The wire coming from the MAF connects to the wire
with label 'From MAF' and the wire that goes to the ECU is connected to the wire 'To ECU'. The 5V supply voltage of this circuit is taken from the
MAF connector too. It only draws a current of a few mA.
Parts
The transistors drawn are BC548B and BC557B, but you can also use 2N2222 and 2N2907 type transistors or similar small signal types. The
resistors are all 1/4W types (1/8W is okay too). C1 and C2 are a ceramic capacitor. C3 is a 15uF/16V electronic capacitor. Use it only if you use a
Mercedes CDI Pierburg MAF. The capacitor filters the signal which may prevent problems during idle with EGR action. Total cost of parts will
only be a few dollar.
By the way: the circuit is not intended to be used with a Pierburg TDI MAF (A3 hot wire MAF). I don't know how a failing A3 MAF behaves, so
the booster circuit may not help in that situation.
If you like to use it, make sure you place it in a water tight housing. Otherwise corrosion will occur. A 35mm film container will do. Stick the wires
through small holes and seal it with silicon.
Intellectual property
You are free to use the information presented here. I do not ask any money for it, but also do not accept any liability. You are not allowed to sell the
schematics or printed circuit boards with a profit. If you find the circuit valuable and are going to use it, I would be happy to hear your experience
with it.
Page Last updated:
The intake pressure and temperature sensor is located just after the intercooler. The sensor has 4 pins.The VW number is 038 906 051 and Bosch
number 0 281 002 177 for a 66kW and 85KW TDI. The wire colors and the connections mentioned to the ECU are from a 66kW ALH or AGR and
a 81kW AHF engine.
Pin
Description
Wire color
Connected to
Signal ground
Brown/blue
Grey/green
Purple/red
Yellow/black
The intake temperature sensor is a simple NTC resistor connected between pin 1 and 2. The resistance is 2000 ohms at 25 degrees centigrade.
Measurements indicate that inside the ECU a 730 ohms resistor is placed, connected to the 5V supply on one side and to the NTC on the other.
The coolant temperature sensor are actually two sensors: one for the engine control unit and one for the instruments control unit. The sensors are
both simple NTC (negative temperature coefficient) resistors. The ECU sensor has a constant signal, the other sensor is sampled.
Pin
Description
Connected to
Ground*
(*: low resistance to ground)
Brown/Green
(Brown/blue)
Pin
Description
Connected to
Brown/Blue
(Brown/yellow)
Ground
Brown White
(Brown)
ground
Purple
It is calculated that there is a 730 ohm resistor connected between 5V and pin 70 in the ECU.
The instruments computer reads the coolant temperature sensor 40 times a second. The voltage on the sensor is high for about 4-6 msec every time
it is read. Therefore the sensor voltage can not be measured with a normal voltage meter. One would need an oscilloscoop or a (fast) voltage meter
with a maximum voltage function. It is calculated that there is a 230 ohm resistor connected between 5V and pin 8 in the instruments computer.
To check the sensors one can easily measure the resistance with an multi-meter. The resistance measured should be according to the figure below.
Note that the resistance has a log-scale. As the instruments ECU doesn't report values below 30 deg.C., the curve stops here. The curve can be
extended, so at 0 deg.C. it is expected that the instruments NTC is about 2200 ohms.
Example: If the engine is 90 deg.C. the instruments NTC should read about 115 ohms and the engine NTC should read about 250 ohms.
Page last updated:09/25/2004
How is it connected
The output wire of the MAF (pin 5 of Bosch MAF and pin 6 of Pierburg MAF) is cut and the circuit below is inserted. The wire coming from the
MAF connects to the wire with label 'MAF Output' and the wire that goes to the ECU is connected to the wire 'To ECU'. The supply voltages of this
circuit are taken from the MAF connector too. It only draws a current of a few mA. One more wire is connected to the EGR. Wire colors may differ,
but it mostly is red/yellow. The other wire (black/yellow) of the EGR is connected to the infamous relay 109, it's the 12V supply. This circuit
doesn't need this black/yellow wire. The flip switch X1 is set to the upper position to enable the circuit. If the circuit is enabled, the EGR valve
should stay closed. If it is stuck open, close it by hand and plug the vacuum hose so it doesn't open again. Do not disconnect the EGR wires, the
ECU will again force limp mode. If the EGR coil is burned, connect a 100 ohm 2W resistor across the EGR wires. The lower switch position is used
to return to normal operation.
A VW Bosch TDI MAF sensor (06A 906 461), which is believed to be good, is used as a reference. The air flow is measured with this sensor while
the output voltage of the CDI MAF is measured.
Clearly the difference can be seen between the two sensors. The original VW sensor outputs a slightly higher output voltage, which will produce a
higher airflow reading in the ECU. At 350kg/hour the curves seem to be closer than at 200kg/hour, but still the TDI ECU will calculate a difference
of about 25kg/hour.(7% difference)
Also remember that the readings above 200kg/hour are done on the road. These will be less accurate as conditions will never be perfectly the same.
The output voltage of the CDI MAF is calculated by reading the RPM and actual MAF value and by the use of the known Bosch TDI MAF curve.
The picture below shows a typical VAGCOM logging results (Fourth gear, WOT). One is with a CDI MAF, the other with a TDI MAF. The time
scale is corrected (time shifting) for one logging to get a good comparison.
Two things are clear (I think).
1. The MB CDI MAF outputs a lower reading. This is in agreement with the curve above.
2. As RPMs climb at the same rate, there's no significant difference in engine power in this condition.
During normal driving conditions the ECU will compensate for the lower MAF output voltage of the CDI MAF. If found that the ECU closed the
EGR by an extra 10% at 120km/hr (75mph). This way the ECU regulates the actual MAF value to the expected value. So the engine gets more fresh
air and less recirculated exhaust gasses. As this will result in more NOx, this may be an environmental issue. So beware!
So how good is it?
It is pretty close, and I cannot tell the difference when I'm driving. (81kW TDI) The readings are not perfect of course, but it is much better to use a
MB CDI MAF in a TDI than a VW 2.0L MAF. The 2.0L MAF produces really high reading which may result in smoke.
First thing to notice is that the CDI MAF will drop its output signal all the way to 1V when the EGR is active during idle. 1V indicates that the
airflow is zero at that moment. Also the amplitude of the signal is larger than the TDI MAF. The average signal is slightly lower (as expected).
Secondly the CDI output signal is clearly made by a digital to analog converter. Notice the steps in the signal. So most likely the analog signal from
the actual sensor element is digitalized, maybe digitally trimmed to get the calibration right, and after that converted to an analog signal again. I
don't think the ECU will mind the steps in the signal. But as some people are getting fault codes while using this MAF, the ECU may not like the
signal drops to 1V. Maybe the signal drops can be avoided by doing the EGR adjustment as is discussed on the webpages of www.tdiclub.com. I
may try it later.
By the way: It is fully normal that the MAF signal is not a steady line: the airflow going into the engine isn't! When a cilinder is breathing air, the
airflow is up. When all valves are closed, the airflow will drop.
To test the MAFs while driving I measured simultaneous both MAFs. One MAF is connected to the ECU, while both are connected to an analog
logging system.
It's difficult to draw a good conclusion as it seems that the airflow is dependent on the position of the MAF sensor. Although of course the total
airflow is the same, the uniformity of the airflow isn't. It seems that the mass air flow graph above is accurate, but above 250 kg/hr the curves may
cross again.
Page last updated: Jan 12, 2005
J1962 Connector
If you can get a J1962 connector for a reasonable price, don't read any further. Just buy one. If not, continue reading.
A J1962 connector is about 15 to 25 euro at the places I looked. To much I think (I'm cheap). So in stead I bought a female connector at the VW
dealer for 4 euro and changed it into a male.
I got part number 3 A0 972 695 from my dealer and stuck a few pins in it:
After that I glued a few plastic parts on it and connected it to my OBDII interface: